Computational and Experimental Studies of Jet Fuel …leads to the fragmentation of heavier alkanes...
Transcript of Computational and Experimental Studies of Jet Fuel …leads to the fragmentation of heavier alkanes...
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Computational and ExperimentalStudies of Jet Fuel Combustion
Mitchell D. Smooke and Alessandro GomezDepartment of Mechanical Engineering
Yale UniversityNew Haven, CT 06520-8284
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Objectives• Examine flame structure of JP-8 doped
flames• Compare and contrast with promising
surrogate candidates• Model the surrogate(s) with semi-
detailed chemical kinetics• Reduce chemical kinetics scheme
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Experimental System• One-dimensional counterflow
flame– complex kinetics but simple
fluids• Liquid dispersion by the
electrospray, followed by sprayevaporation in the fuel line
• Perturbed baseline methaneflame– Use flame as a “controlled”
reactor by maintaining a fixedtime-temperature baseline
– Avoid condensation problemsand “cooking” the fuel in thetransfer lines
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
• 30µm tip size 100µm body (but 3Xprobe maintains comparable spatialresolution at atmospheric pressure)
• SiO2 for high temperature tolerance
• Small sampling flow rate (≈ 50µ l/min)
Microprobe Details
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
4 hours of flame burning and …..30 hours of off-line analysis!
GC/MS Configuration
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Experimental Conditions (non-Sooting Flames)• Baseline Flame
– Fuel stream XN2 = 0.935 XCH4= 0.065
– Oxidizer stream XN2 = 0.74 XO2= 0.26
• Dopant perturbation
– 4.2 103 ppm of either JP-8 or surrogate
• Max. Temp= 1750K
• Strain Rate= 102 s-1
• Zf (stoich. mixture fraction)= 0.77
• Flame on fuel side of stagnation plane
Non-sootingGSP
Fuel
Oxidizer
Flame
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Numerical Simulations•1D Simulation using self-similar transformation
•Dirichlet conditions imposed at the inlet
•221 Species × 5032 reactions
[E.Ranzi et al., Prog.Energ.Comb.Sci, 2001]
•Pseudo-time stepping + Newton’s Method
Experimental results have been shifted by 1.0mm to accountfor probe intrusiveness and other effects
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
From [Violi et al., Comb. Sci. Tech., 2002], designed to matchdistillation properties, smoke point behavior, laminar flame speedand validated for extinction limits in (Cooke et al., 2005)
• 10% iso-Octane
• 30% n-Dodecane
• 20% n-Tetradecane
• 20% Methylcyclohexane
• 15% o-Xylene
• 5% Tetralin
Surrogate Formulation
……but ready to test other promising alternatives (e.g., Aachen, SERDP, MURI, etc.)
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Results• Measured and computed mole
fractions for
• Large alkanes
• Ethane, ethylene, acetylene,methane
• Major species (CO, CO2 andO2)
• Aromatics
• Temperature measurementscorrected for radiation losses
• Zoom in on fuel side of the domain
• x = distance from fuel inlet
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Major Species and TemperatureJP-8 doped-
Symbols/continuous line Surrogate-doped (exp)-
Symbols/dashed line Surrogate-doped (comp)-
Continuous thick line
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
JP-8 vs. Surrogate: Large Alkanes
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Surrogate- Experiments vs. Computation: Large Alkanes
Surrogate-doped (exp)-Symbols/dashed line
Surrogate-doped (comp)- Continuous thick line
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
JP-8 vs. Surrogate: C4-C6 Species
JP-8 doped- Symbols/continuous line
Surrogate-doped (exp)- Symbols/dashed line
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
C2 species
JP-8 doped- Symbols/ continuous line
Surrogate-doped (exp)- Symbols/ dashed line
Surrogate-doped (comp)- Continuous thick line
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Aromatics
Surrogate-doped (exp)-Symbols/dashed line
Surrogate-doped (comp)- Continuous thick line
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Aromatics (close-up)
JP-8 doped- Symbols/ continuous line
Surrogate-doped (exp)- Symbols/ dashed line
Surrogate-doped (comp)- Continuous thick line
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Conclusions (Part 1)• The addition of jet fuel to a nonsooting baseline methane flame
leads to the fragmentation of heavier alkanes to smaller ones,down to C2-hydrocarbons and the appearance of peak aromaticconcentrations surviving further into the high temperatureregion.
• A 6-component surrogate captures the pyrolysis and oxidationbehavior of JP-8 reasonably well, with the most significantdiscrepancy for benzene, toluene and ethylene.
• The computational results using a semi-detailed kineticmechanism are in reasonably good agreement with theexperiments.
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
More flames?
With what feed streamcomposition?
Objective: span a broad range ofZf, from non-sootingflames to sooting ones
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
• Max. Temp= 1750K
• Strain Rate= 102 s-1
• Zf (stoich. mixture fraction)= 0.77
• Flame on fuel side of stagnationplane
Non-sooting Flames
GSP
Fuel
Oxidizer
Flame
• ↑ Max. Temp
• ↓ Strain Rate
• ↓ Zf
• Flame on oxidizer side of stagnationplane
Sooting Flames
GSP
Fuel
Oxidizer
Flame
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
2. Experimental Conditions (Incipiently Sooting Flames)
GSP
Fuel
Oxidizer
Flame
• Baseline Flame #2
– Fuel stream XN2= 0.728XC2H4= 0.272
– Oxidizer stream XN2= 0.807XO2= 0.184
• Dopant perturbation
– 2×103 ppm of either JP-8 or surrogate
• Max. Temp= 1989 K
• Strain Rate= 67.8 s-1
• Zf = 0.18
• Flame on oxidizer side of stagnationplane
≈0.6mm
Sooting
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Major Species and Temperature
0.00
0.05
0.10
0.15
0.20
0.25
0.30
0 1 2 3 4 5 6 7 8 9 10 11
Distance from fuel inlet [mm]
Mo
le f
rac
tio
n
300
700
1100
1500
1900
2300
Ethylene
CO2
CO
Temperature [K]
JP-8 doped- Symbols/ continuous line
Surrogate-doped (exp)- Symbols/ dashed line
Baseline (exp)- small dash line
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Surrogate Components
0
100
200
300
400
500
600
0 1 2 3 4 5 6 7
Distance from fuel inlet [mm]
pp
m
Methylcyclohexane
Isooctane
Dodecane
Tetradecane
Tetraline
Xylene
0
10
20
30
40
50
60
70
80
0 1 2 3 4 5 6 7
Distance from fuel inlet [mm]p
pm
Meth.Cyc.Hex
Heptane
Isooctane
Octane
Nonane
Decane
Undecane
Dodecane
Tridecane
Tetradecane
Pentadecane
JP-8 Alkanes
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
C1 and C2 Species
0.0E+00
1.0E-04
2.0E-04
3.0E-04
4.0E-04
5.0E-04
6.0E-04
7.0E-04
8.0E-04
9.0E-04
1.0E-03
0 1 2 3 4 5 6 7 8 9
Distance from fuel inlet [mm]
Mo
le f
racti
on
0.000
0.002
0.004
0.006
0.008
0.010
0.012
Ethane
Methane
Acetylene
JP-8 doped- Symbols/ continuous line
Surrogate-doped (exp)- Symbols/ dashed line
Baseline (exp)- small dash line
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Aromatics
0
20
40
60
80
100
120
140
160
1 2 3 4 5 6 7 8
Distance from fuel inlet [mm]
pp
m
Benzene
Toluene
JP-8 doped- Symbols/ continuous line
Surrogate-doped (exp)- Symbols/ dashed line
Baseline (exp)- small dash line
Soot band
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Chromatogram Comparison
• Grassy background
• Overlapping peaks
• 30 species identified and measured
• 20 species identified but notquantified
•Clean distinct peaks
•Possibility to track minor species
Surrogate JP-8
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Carbon Count Problem
Methane baseline flames (exp 1)CO and CO2 excluded from count
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Perturbation of Sampling Probe• The objective is to assess the probe perturbation
in terms of flame displacement by OH PLIF• A potentially useful byproduct is the concurrent
PLIF from PAH blends in the JP-8
PAH PLIF
OH PLIF
Nd:YAG(532 nm, ~60 mJ, 6 ns)
Dye Laser(566 nm, ~25 mJ)
BBO (x2)(~283 nm, ~4 mJ)
ICCDcamera
Flash LampTrigger Output
UG5 + WG305 Filter Set(52.1% Transmittance
@ 310 nm)
PIV Laser
PIVcamera
532 nm IF
Telescopeoptics
Burner
Energy meter
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Probe position above fuel
outlet
OH PLIF image
2.8 mm
3.8 mm
4.8 mm
5.8 mm
6.8 mm
7.8 mm
8.8 mm
9.8 mm
Flame shift ≤ 0.6 mm
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Conclusions (Part 2)• Experiments on an incipiently sooting ethylene flame show good
agreement between surrogate and JP-8 even with respect tobenzene and toluene, suggesting that the surrogate maycapture even the sooting behavior of JP-8, as intended by thematching of the smoke point in the surrogate design.
• A procedure was demonstrated to reduce the experimental timeto operate the burner for a complete flame scan toapproximately 4 hours, with 30 hours of automated off-lineGC/MS/FID chemical analysis.
• The JP-8 chemical analysis is invariably incomplete andaccounts for roughly only 20% of the overall carbon introducedas liquid fuel.
• OH PLIF showed a modest perturbation by the sampling probe.
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Future Work (pending renewal)
• Examine systematically other surrogate candidatesand their individual components (e.g., Aachen, MURI,SERDP);
• Tweak the chemical kinetic mechanism in thecomputational model to improve agreement withexperiments (in collaboration with Ranzi’s group inMilan);
• Use sensitivity tests in the chemical kinetic model forguidance in the chemical analysis;
• Design and test a high pressure chamber to performsimilar work at high pressures (up to 40 atm);
• Implement a reduced kinetics strategy.
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
References• Tosatto et al, to appear in the Proceedings of the 32nd International
Combustion Symposium, 2008• Bufferand et al., submitted to C&F, 2008• Jahangirian et al., in preparation, 2008
Acknowledgements
• Graduate students Luca Tosatto, Hugo Bufferand and Saeed Jahangirianperformed the bulk of the experiments and Bruno Coriton helped with theOH LIF experiment.
• Dr. James (Tim) Edwards (WPAFB) supplied the JP-8.• Research supported by
– AFOSR (Grant #FA9550-06-1-0018, Dr. Julian Tishkoff, Program Manager)and
– ARO DURIP (Grant # W911NF-07-1-0231, Dr. Ralph Antonien, ProgramManager)
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Back up
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
From JP8: aliphaticstoluene,xylene,ethylbenzene,n-alkylbenzenes,etc.
isopropylbenzeneetc.
benzene benzyl styrene
naphthalene
(toluene )
sootCourtesy of Charles Mc Enally
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
C2 Hydrocarbons in a JP-8-doped hydrogen flame (fullsymbols) and a surrogat-doped hydrogen flame (open
symbols).
Fuels Summit, NIST, Gaithersburg, Maryland, September 8-10, 2008
Flame A Flame B Flame C Flame D
Molar Composition
N2
C2H4
C2-C5 alkane (Ethane)
impurities
0.7278
0.2722
851 ppm
0.7340
0.2641
825 ppm
0.7339
0.2641
825 ppm
0.7339
0.2641
825 ppm
JP-8 (C11H21)
1953 ppm
Utah Surrogate
Methyl-Cyclohexane
Iso-Octane
m-Xylene
Tetraline
Dodecane
Tetradecane
Total=
394 ppm
197 ppm
295 ppm
98 ppm
591 ppm
394 ppm
1970 ppm
Aachen Surrogate
1,2,4-trimethylbenzene
n-Decane
Total=
450 ppm
1520 ppm
1970 ppm
Mass Flux (g/min/cm2) 1.619 1.684 1.683 1.682
Fuel
Side
Temperature (K) 417
Molar Composition
N2
O2
0.8070
0.1843
Mass Flux (g/min/cm2) 1.891 1.925 1.925 1.925
Oxidizer
Side
Temperature (K) 414
Strain Rate (s-1
) 67.78 69.54 69.53 69.52
zf 0.1798 0.1797 0.1799 0.1801
Sooting Flames B.C.
In the surrogate-doped flame,B, 14.2 103 ppm of C as C2H4in the baseline flame, A, arereplaced with 11.0 103 ppm ofC as JP-8